近年來由於無人飛機發展迅速而衍生出許多議題，其中防撞的機制更是最讓人關注的，有過無人機飛入民宅及衝撞大廈導致障礙物損壞或是人員受傷，因此本研究將發展一套防撞系統應用在四旋翼機上，當感測器偵測到障礙物時會進行迴避的動作。
本研究的目的是建立起一套決策樹防撞系統的四旋翼機，使四旋翼機距離障礙物太近時會自動迴避的功能。當超音波測距儀所得到的距離值被判斷為防撞區間，防撞系統控制權限將接手控制載具姿態，迴避之俯仰姿態角度乃根據操作者實際操作四旋翼機所得到的經驗而定，執行以下三種迴避:初步緩速迴避(Case 1)、緩速迴避(Case 2)以及急速迴避(Case 3)，而防撞的飛行數據將藉由無線傳輸模組，即時傳至地面資料站進行進一步的分析與討論。
整體決策樹防撞演算法、傳輸及地面資料站的系統已於實際防撞飛行測試中得到驗證。嵌入的防撞演算法控制系統能夠穩定飛行，於防撞區域時迴避系統控制的權限比操作者控制權限高，進而執行迴避程序，且可以在約為每秒1.9公尺的飛行速度下成功遠離障礙物。

The purpose of the research is to build a decision tree algorithm on collision avoidance system used for quadcopters. While the ultrasonic range finder judges the distance is in collision avoidance interval, the access will be replaced from human to the system which operates attitude of the UAV. According to the former experiences on operating quadcopters, it is obtained the appropriate pitch angle. The UAS implement the following three motions to avoid collisions. Case(1): initial slow avoidance stage, Case(2): slow avoidance stage and Case(3): rapid avoidance stage. Then the training data about collision avoidance test will be transmitted to the ground station via wireless transmission module to further analysis. The entire decision tree algorithm on collision avoidance system, transmission data, and ground station have been verified in flight tests. During the test, the quadcopter can implement avoidance motion in real-time and move away from obstacles steadily. In the avoidance area, the authority of the collision avoidance system is higher than the operator and implements the avoidance process. It is successful to fly away from the obstacles in the 1.92 meter per second and the minimum distance is 1.05 meters.

[1]Amit K. Tripathi, Ramsingh G. Raja, Radhakant Padhi, “Reactive Collision Avoidance of UAVs with Stereovision Camera Sensors Using UKF,” IFAC Proceedings Volumes, vol.47, pp.1119-1125, 2014.
[2]K.Y. Chee, Z.W. Zhong, “Control, Navigation and Collision Avoidance for an Unmanned Aerial Vehicle,” Sensors and Actuators A:Physical, vol190, pp.66-76, 2013.
[3]Yongsoon Yoon, Tokson Choe, Yongwoon Park, “Obstacle Avoidance for Wheeled Robots in Unknown Environments Using Model Predictive Control,” IFAC Proceedings Volumes, vol.41, pp.6792-6797, 2008.
[4]Aydın Eresen, Nevrez İmamoğlu, Mehmet Önder Efe, “Autonomous Quadrotor Flight with Vision-Based Obstacle Avoidance in Virtual Environment,” Expert Systems with Applications, vol.39, pp.894-905, 2012.
[5]蕭仲仁, “應用模糊決策樹控制法則於自走車防撞系統之研發,” 國立成功大學航空太空工程學系碩士論文, 2009.
[6]吳秉勳, “四旋翼防撞系統設計與驗證研究,” 國立成功大學航空太空工程學系碩士論文, 2015.
[7]Decision tree, https://goo.gl/LEyjMd
[8]Huaining Sun, Xuegang Hu, “Attribute Selection for Decision Tree Learning with Class Constraint,” Chemometrics and Intelligent Laboratory System, vol.163, pp.16-23, 2017.
[9]Daniel Westreich, Justin Lessler, Michele Jonsson Funk, “Propensity Score Estimation: Neural Networks, Support Vector Machines, Decision Trees (CART), and Meta-Classifiers as Alternatives to Logistic Regression,” Journal of Clinical Epidemiology, vol.63, pp.826-833, 2010.
[10]Michael J. Moran, Howard N. Shapiro, Fundamentals of Engineering Thermodynamics, 文景書局有限公司, 1993.
[11]I2C-Bus Specification and User Manual, https://goo.gl/fRqImJ
[12]ArduPilot Autopilot Suite, http://ardupilot.org/ardupilot/index.html
[13]MaxBotix High Performance Ultrasonic Rangefinders MB1202 Datasheet, https://goo.gl/vvxuTX
[14]高橋麻奈, C++物件導向程式設計實例入門, 博碩文化股份有限公司。